Calibrating Selling and Loss Intensities for Liquidity Stress Tests

Summary

The document asks how to interpret and calibrate selling intensity and loss intensity in a liquidity stress test for a portfolio of equities and bonds. It introduces a cash conversion factor that combines a liquidity factor, based on selling intensity over time, with a discount factor, based on loss intensity scaled by the square root of time and capped by maximum drawdown. The cited paper supplies example intensity values, but the document does not explain how they were estimated or provide a calibration procedure.

The central issue is how to choose asset-specific inputs that reflect how quickly holdings can be sold and how much value they may lose while being liquidated during stress. The post itself offers no answer, data, or empirical results; it is a request for guidance from someone with limited statistical experience. Any practical calibration would therefore need supporting assumptions and evidence beyond what appears here, including a clear stress horizon and relevant market data.

Key ideas

• The proposed cash conversion factor combines a sale-rate component with a loss component.
• Selling intensity controls how quickly the modeled position can be converted to cash.
• Loss intensity scales the modeled discount with the square root of time, subject to a maximum drawdown cap.
• The document asks how to calibrate the inputs for equity and bond holdings but provides no calibration method or supporting evidence.

Tags

# What is selling intensity, loss intensity, and how can I calibrate them?


# What is selling intensity, loss intensity, and how can I calibrate them?












Thought asking around on a problem I'm currently facing. I have a hypothetical multi-asset portfolio of equities and bonds, on which I'm trying to measure it's liquidity risk in stressed periods. I've read Roncalli's four-piece paper on Liquidity Stress Testing in the Asset Management Industry (link), and in it they suggest using a cash conversion factor for a sixty day period, to determine the price impact on the asset over that period. It is as follows:

Cash Conversion Factor(time) = Liquidity Factor x (1 - Discount Factor), where Liquidity Factor = Min(1, selling intensity x time); Discount Factor = Min(Maximum Drawdown, Loss Intensity x square root(time)).

In it, they suggest a selling intensity of 5% and a loss intensity of 6.25%, which would result in the following

However, I'm unclear on what they mean about selling and loss intensities, much less how it is derived, and how can it be calibrated, for bonds and equities. Can anyone assist? And apologies if it sounds too simplistic, I only have a basic understanding of econometrics and statistics!

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.